Forschungsschwerpunkte
Anwendung von maschinellem Lernen (ML) zur datengetriebenen Alterungsmodellierung in der PEM-Wasserelektrolyse
- Rein datengetriebene Alterungsvorhersage mittels Deep Learning
- Physics-informed ML zur Steigerung der Vorhersagegenauigkeit und Interpretierbarkeit
- Untersuchung von Explainable AI (XAI)-Ansätzen zur Identifikation von Alterungsstressoren
Werdegang
| seit 2021 | Wissenschaftlicher Mitarbeiter am Institut für elektrische Energiesysteme an der Leibniz Universität Hannover |
| 2018-2021 | Studium M. Sc. Maschinenbau, Leibniz Universität Hannover
|
| 2013-2018 | Studium B. Sc. Maschinenbau, Leibniz Universität Hannover
|
Publikationen und Projekte
-
Publikationen (Hier klicken)
Showing results 1 - 3 out of 3
Predicting future polarization curves from operating data: Machine learning-based investigation of degradation modeling concepts for PEM water electrolysis. / Woelke, Janis; Rex, Alexander; Eckert, Christoph et al.
In: Energy and AI, Vol. 21, 100547, 09.2025.Research output: Contribution to journal › Article › Research › peer review
Cross-Method Overview of Fleet-Based Machine Health Estimation and Prediction: A Practical Guide for Industrial Applications. / Yan, Xuqian; Woelke, Janis; Bensmann, Boris et al.
In: IEEE ACCESS, Vol. 13, 10.04.2025, p. 60131-60147.Research output: Contribution to journal › Review article › Research › peer review
Machine learning in proton exchange membrane water electrolysis: A knowledge-integrated framework. / Chen, Xia; Rex, Alexander; Woelke, Janis et al.
In: Applied energy, Vol. 371, 123550, 01.10.2024.Research output: Contribution to journal › Article › Research › peer review
-
Forschungsprojekte (Hier klicken)
Showing results 1 - 1 out of 1
SEGIWA: FleetTwin-LUH
Hanke-Rauschenbach, R. (Principal Investigator)
1 May 2021 → 30 Sept 2025
Project: Research
Abstract: Siemens Energy develops, builds and markets PEM-based water electrolyzers. In SEGIWA, basic principles are to be developed to transfer the series Silyzers300 series from manual production to series production in the direction of gigawatt expansion. In line with the national hydrogen strategy, the aim is to achieve a low-friction market ramp-up. Concepts for scaling electrolyzer production on an industrial scale support this ramp-up. The academic partners are developing methods for process control and quality assurance. Alternatives to catalysts and membranes are being considered and possibly transferred to series production. MES, automation, production-accompanying online analytics and the digital twin form core work packages.
The subject of the present sub-project are the contributions of the partner LUH to the SEGIWA work package 4 "De-risking through digital integration of the value chain". The main focus of the planned work is on the scientific support of the development of the so-called FleetTwin, a digital twin, with the help of which the continuous analysis of the operating data of the numerous customer plants created in the context of series production should be possible. In this way a feedback loop can be established from the field into production as well as into development and engineering.